Server :: Red Hat - Multipathing In Conjunction - Fdisk -l To Display One Lun Rather Than Displaying All 8 Paths To The Lun?
Aug 25, 2010
How to device-mapper multipathing in conjunction with fdisk. I am more of a solaris expert and now trying to grasp linux (red hat). I have a server that I plan to cluster (future assignment) which is attached to a Netapp SAN via emulex LPe11002 HBA. I have 8 paths to the LUN I created and I can see the multipath device (/dev/mapper/mpath1). Below is my multipath.conf file:
[Code]....
Question 1: When I look at fdisk -l, I can see all 8 paths to the same LUN. Although I partition /dev/mapper/mpath1, I can still see that 7 of the 8 still say "Disk /dev/sdX doesn't contain a valid partition table". I would think that all 8 would have this removed. I execute '/usr/sbin/lpfc/lun_scan all' and it doesn't work. I am looking to clean this up...what should I do? It seems weird to have to do it to all 8 paths!
Question 2: Why does my alias, 'db' not show up under /dev/mappers?
Question 3: Is there a way for fdisk -l to display one lun rather than displaying all 8 paths to the lun?
What is the standard procedure (commands) people execute when adding a new lun to a server with multipathing so it is nice and clean?
I'm a storage engineer by trade and have mapped a RHEL 4 server, several LUNs. I've mapped each (HBA Initiator) 4 times to seperate fibre channel ports on a storage array.Is there a RH command to see the device mapped 4 times to the target. I would expect to see something in the output that has the subsystems WWN, such as -
Multipathing in linux -need to check how the luns are configuredI have no idea about luns configuration in linux environment(newbie) and now i have issue figuring out the configurationSenario,By the look of the multipath output 3 luns are connected to server1 is from Hitachi and 2 from Sunstorage(Thumper)What i am trying to achieve,I need to check how the 3 luns are configured on OS levelI need to differentiate Hitachi and thumper lun The next is to check whether 3 luns are mirrored or not (if not ,how they configured)The third and last .I need to remove the mirroring luns (if exist) and leave the system with only Hitachi lun
Has anyone seen or heard of the following problem? I have a 1in black border surrounding my display. I've been running google searches all day without much luck. My current setup is as follows: Ubuntu 9.10 (64-bit), 24in asus lcd, Radeon 5770, amd phenom 2 x4, ddr3 4gb, gigabyte mb. I noticed the border while running the live cd. I figured it had something to do with the video drivers. Last night I installed ATI's linux driver 10.2. I'm getting my preferred resolution of 1920x1080 over dvi. But i'm not able to use my full 24in. It's like using a 20in lcd.
I have three boxes with lenny debian recently installed. I'd like to be able to display X apps running on host1 and displaying on host2. "xhost +" would do the trick for me if I could get it to work but as far as I can tell it does nothing. (BTW, I'm aware of the security risks. This is not a concern for me. I'm safely behind a good firewall. I'm also aware of "ssh -X" but that won't quite solve my problem. See the long version below.) I suspect that Debian secures the xhost command out of the box in some way but I can't seem to figure out what it is. If that is true and someone can show me how to reset it, or direct me to appropriate documentation I think that would solve my problem.
Here's the long version:
What I really want to do is play xbattle with my kids but I need to be able to display the board to three different screens. In xbattle the command line is: % xbattle -red me -blue host1:0 -green host2:0 {and more stuff of no consequence to my question}
This opens a game board for player red locally, player blue on host1 and player green on host2. Or at least it should. If there is a way to make this work with "ssh -X" (or anything else for that matter) I'll be happy to use that instead, but so far as I can tell "ssh -X" only lets me redirect the display back to the one I ssh'd from whereas I need to direct to multiple displays to run xbattle.
As many developers probably do, I have a Windows based machine on which I run XAMPP locally to test my code and a Linux machine with Fedora as my remote server.As I sometimes use .htaccess as a way to authenticate some parts of the website, I end up having two .htaccess files: one with the local path (something line D:My_Webs) and one with my remote path (something like /var/www/html/) to the password file.I have searched high and low, but I cannot seem to find any trick so that I only have to maintain one version of the .htaccess file which can work on both Linux and on Windows machines.
I'm relatively new to linux in this capacity. I've had to reboot a SAN host (iSCSI initiator). I took a grab of the df -h output before reboot to ensure I was all mounted again afterwards.
Is there a way to use the grep command in conjunction with an editor such as nano or vi so that I can remove the commented out lines from a conf file and then proceed to edit it ? I can use grep -v "^#^" squid.conf (example ) which gives me a nicely tidy conf file but I can't edit it.Can this command be used with nano or ?
I've got an older RHEL4 system (2.6.9-34.ELsmp) which I am needing to migrate off of one LUN and onto another. My plan was to simply have the storage team assign me a new LUN and to setup LVM mirroring then remove the origional LUN. My problem is with the new LUN. I had to reboot to get the system to even see it. It now shows up in /proc/scsi/scsi but there is only one new device in /dev and I cannot even open the sd* device.. I can only open the dm-* device.
Is there anything special I need to do to use this card in conjunction with Fedora 14? It's been acting a little strange lately. Also I think the card may have some type of 3D support although I'm not entirely sure on that.
We're trying to take an existing system running multipathed SAN boot disks, duplicate that boot disk to another system volume, and boot a second system up with all the same parts running like the first system (think: cloning the system).However, multipathing is broke on the second system 'm sure I'm forgetting something stupid here after thecloning).Details:We have a freshly installed RHEL6 on IBM PPCThis is a boot off of a SAN volume with two paths. Installation went great - multipathing was auto-detected and used underneath the usual LVMs for the boot volumes. Looks and works great:
Code: [root@goldimage dev]# df -k Filesystem 1K-blocks Used Available Use% Mounted on
I recently expanded the RAID on an iSCSI device which is shared out to a linux server. Fdisk correctly sees the new size but it won't let me add a third partition. It complains about overlapping partitions whenever I try to add it.If I 'remove' the partition, the overlapping error goes away. The interesting thing here is that when attempting to use the default sizes to setup the partition the printed screen shows completely different results (which I can understand is why its complaining about overlapping)
I'm not sure why its not accepting the cylinder sizes I want. Anyone have any ideas as to what's wrong here?
I created a new disk on our scsi san. I then ran the following command: echo "- - -" > /sys/class/scsi_host/host1/scan. did that command for each host. and in dmesg, it shows it found a device /dev/sdg. but when i do a fdisk -l. It never lists /dev/sdg. I just did this other day on another server and it worked fine like that. This is RH 4.8.
I install centos 5.4 on my intel S5000vsa motherboard, and with embeded RAID1. I successfully config RAID1 in BIOS, and then I set to install centos use linux dd mode. And also successfully loaded raid drivers. In centos partition step, one disk appears. However, when centos installed and entered the system, strange result in fdisk -l.
It's been a while since I configured a raid and have been making some changes to my main workstation/server. fdisk does not like md devices on my machine... always says it has an invalid partition table. While this is said to be normal all over the net, I don't feel warm and fuzzy about that fact. What is best practice these days, to create a non-partitionable md device or a partitionable mdp device?
If I create a partitionable md device, I would imagine it would look good in fdisk. However, I am concerned about growing the array afterward. I would then have to grow the array, redefine the partition, and then grow the file system. The PITA factor goes up. Has anyone worked with both? Pro/Cons? My array was created with:
I'm using the "Beginning PHP5, Apache, MySQL Web Development" book from WROX to teach myself these processes. I'm a retired programmer and trying to keep my brain active. I've been able to get through most of the process and get things to work. But getting a picture upload and then display it is killing me. The picture is moved from the initial directory and placed in the destination but will not show on the web page, I get a place holder and all the information about the picture, but no pic. The MySQL is processing all the database stuff .I haven't a clue what is not working, Firefox, Apache or what. I'm running Fedora 14.
I have installed Samba and the Guest only share i've created works, I can connect and view, edit, create, delete files fine... However...I can NOT see or create subfolders.Google searches and playing around for six odd hours and no success...
----------smb.conf-------------- # This is the main Samba configuration file. You should read the # smb.conf(5) manual page in order to understand the options listed
# service httpd stop Stopping httpd: [ OK ] And now we access the web site by typing the URL: http://192.168.56.10/ and the following message, as expected, is displayed:
Code:
This webpage is not available. The webpage at http://192.168.56.10/ might be temporarily down or it may have moved permanently to a new web address. But is there any way to display a different web page or message when we have brought down the Apache Server for maintenance so that a visitor on our web site would not think that we are gone with our bags and baggages...?
I have a png file which I put into a user's directory: /~user/file.png This displays OK. When placed in the user's images directory: /~user/images/file.png the file does not display. If I put the file in the global images directory: /images/file.png the file displays OK. Finally if I change the format of the file to a jpg and save it in the user's images directory: /~user/images/file.jpg the file displays OK.
When I look at the Apache access_log and error_log files, there is not even an entry if the png image file is accessed from the user's images directory.
When i ssh to server using -X, i always confuse about which display number i should export. It seems to me sometimes the display number has been used by something, so what i can do is only
I have an Ubuntu 9.10 server set up at my house. I have Apache2 and PHP5 installed on it. Every time I go to the server on a web page and try to load the PHP index page it downloads instead of displaying.
I have virtual servers set up and have the files stored at /home/cusinndzl. If anyone needs to take a look I can let them into the webmin panel.
I have a weird problem I am experiencing with my Apache 2.2 w/ ISPConfig 3 running on Debian 5.0.4. It is not a website html issue.***PROBLEM***Images are not beind displayed,javascripts are not being run, however when the site is viewed from a local source as a file(like on a USB Flash Drive) they work fine and display properly.Here is the website URL:I have removed a lot of the images and content or changed it in order to troubleshoot it myself, so the site is stripped pretty bare. There is supposed to be a logo inside the "header" <div>, which is set in the CSS file "style.css". There are supposed to be images for navigation, but they show up as red "x"'s. I removed the javascript elements and just trying to work with barebone html/css and can't seem to get it to display properly.
I also have an application installed under /usr/share/myappHow to make a link to the application installed under myapp in to a directory that is already in my env variable?
at present I compile the same code for different systems (cross compilers)I need to call libs and include paths for the different processors. At present I simply comment out the paths not needed
I have not been on in about a month or 2. I have no idea how to list this thread. I am hoping that someone like tex can help out. Being lazy with ubuntu seems to have been badong. Ok, I had 2 physical machines and 3 vms. VM's ran under Vbox-ose via a bridge. OS's are buntu x2 and one centOS box. I installed webmin to make things easy ( i thought ).However, after setting webmin, my I have been randomly loosing PATH's. I mean, one minuet i can run sudo apt-get and the next the whole PATH is gone. I tried a compare of my "home" box's bashrc and bash_profile against the other machins, and outside of some alias for colr and the like nothing seems to stand out. Even if I su - to root I am not getting access to the needed paths. Now, while I could export the correct path, I am more conserned with the why of it all.
I would have thought that as long as my group setting on my ssh users were all correct AND the environment had not been changed, all would be good. I can provide more info if someone wants to help me out with this. HOwever, it drove me to a six pack. 8SI have read the man pages. I have used google. I have checked the logs. The logs by the way showed a lil hammering on one of the boxes for root access. [I]t wasnt me. However, I dont seem to be able to see a time stamp.
Is there some sort of standards file path convention for installing softwares that I could follow through? For example, I just learnt how to build Nginx from source. But the default binary path set by nginx is "/usr/local/nginx/sbin". I have seen a couple of tutorials which they specify the location of the installed binary and it is very different from those usual default paths. Thus, got me thinking whether is there some form of file path convention that I should follow?Is there some kind of list which states where do those packages on Debian.org Repository usually installed to?
Adobe Reader ain't available for 64-bit so I'm stuck with XPDF (or Okular, which is better than XPDF) and most stuff won't print worth a hoot. I've been reading the manual pages and somewhere or other I'm missing some thing about adding font paths (or replacing the default font paths). I have the complete Adobe Type Library plus TTF's installed in /usr/local/share/fonts (and they work just fine with OpenOffice.org) but I can't seem to figure out how to get them or the Slackware-provided fonts available to XPDF.
I'd like to backup my whole system to a 2nd disk using rsync (other tools not possible).Which paths should I exclude from the packup?I was thinking about /proc, /dev, the lost+found directories...What other paths am I forgetting?
